Abstract
Facial expression recognition (FER) can assist the interaction between humans and devices. The combination of FER and ensemble learning can usually improve final recognition results. However, in many cases, the produced ensemble classifiers often contain many redundant members, and those components bring potential side effects to final results. Previous studies have illustrated that a more compact subset of a classifier pool shows better performance than original classifier pool. Furthermore, the compacted subset reduces storage space and decreases computation complexity. This paper proposes a maximum relevance and minimum redundancy-based ensemble pruning (MRMREP) method that treats prediction results as features, and extends the feature selection method to the ensemble classifier reduction problem to obtain a more representative subset. This novel method ordered all base classifiers according to two important factors: the correlation between target labels and predictions, and the redundancy between classifiers. The final ensemble performance was evaluated by comparing our method with other ensemble pruning methods, and superior results were obtained on the FER2013, JAFFE, and CK + databases.
Similar content being viewed by others
References
Abdel-Hamid O, Abdel-Rahman M, Hui J, Deng L, Penn G, Yu D (2014) Convolutional neural networks for speech recognition. IEEE ACM Trans Audio Speech Lang Process 22(10):1533–1545
Cavalcanti GDC, Oliveira LS, Moura TJM, Carvalho GV (2016) Combining diversity measures for ensemble pruning. Pattern Recogn Lett 74:38–45
Chang K-H, Parker DS (2016) Complementary prioritized ensemble selection. In: 2016 international joint conference on neural networks (IJCNN)
Chen T, Su F (2012) Facial expression recognition via Gabor wavelet and structured sparse representation. In: Proceedings—2012 3rd IEEE international conference on network infrastructure and digital content, IC-NIDC 2012, pp 420–424
Cheng J, Liu G-Y (2015) Affective detection based on an imbalanced fuzzy support vector machine. Biomed Signal Process Control 18:118–126
Cruz RMO, Sabourin R, Cavalcanti GDC (2014) On meta-learning for dynamic ensemble selection. In: 2014 22nd international conference on pattern recognition (ICPR), pp 1230–1235
Dai Q (2013) A novel ensemble pruning algorithm based on randomized greedy selective strategy and ballot. Neurocomputing 122:258–265
Dai Q, Han X (2016) An efficient ordering-based ensemble pruning algorithm via dynamic programming. Appl Intell 44(4):816–830
Dai Q, Liu Z (2013) ModenPBT: a modified backtracking ensemble pruning algorithm. Appl Soft Comput 13(11):4292–4302
Dai Q, Zhang T, Liu N (2015) A new reverse reduce-error ensemble pruning algorithm. Appl Soft Comput 28:237–249
Dasarathy BV, Sheela BV (1979) Composite classifier system design: concepts and methodology. Proc IEEE 67(5):708–713
de Souto MCP, Soares RGF, Santana A, Canuto AMP (2008) Empirical comparison of dynamic classifier selection methods based on diversity and accuracy for building ensembles . In: 2008 IEEE international joint conference on neural networks (IEEE world congress on computational intelligence), pp 1480–1487
Diao R, Chao F, Peng T, Snooke N, Shen Q (2014) Feature selection inspired classifier ensemble reduction. IEEE Trans Cybern 44(8):1259–1268
Giannakakis G, Pediaditis M, Manousos D, Kazantzaki E, Chiarugi F, Simos PG, Marias K, Tsiknakis M (2017) Stress and anxiety detection using facial cues from videos. Biomed Signal Process Control 31:89–101
Gong X, Yang Y, Lin J, Li T (2011) Expression detection based on a novel emotion recognition method. Int J Comput Intell Syst 4(1):44–53
Goodfellow IJ, Erhan D, Carrier PL et al (2013) Challenges in representation learning: a report on three machine learning contests. Neural Inf Proces 8228 LNCS (part 3):117–124
Guihua W, Huihui L, Danyang L (2015) An ensemble convolutional echo state networks for facial expression recognition. In: 2015 international conference on affective computing and intelligent interaction (ACII), pp 873–878
Hady MFA, Schwenker F, Palm G (2010) When classifier selection meets information theory: a unifying view. In: 2010 international conference of soft computing and pattern recognition (SoCPaR), pp 314–319
Hamester D, Barros P, Wermter S (2015) Face expression recognition with a 2-channel convolutional neural network. In: 2015 international joint conference on neural networks (IJCNN)
Ijjina EP, Mohan CK (2014) Facial expression recognition using kinect depth sensor and convolutional neural networks. In: 2014 13th international conference on machine learning and applications (ICMLA), pp 392–396
Jia X, Zhang Y, Powers D, Ali Humayra B (2014) Multi-classifier fusion based facial expression recognition approach. KSII Trans Internet Inf Syst 8(1):196–212
Krawczyk B (2015) One-class classifier ensemble pruning and weighting with firefly algorithm. Neurocomputing 150:490–500
Kuncheva LI (2013) A bound on Kappa-Error diagrams for analysis of classifier ensembles. IEEE Trans Knowl Data Eng 25(3):494–501
Kuncheva LI, Whitaker CJ (2003) Measures of diversity in classifier ensembles and their relationship with the ensemble accuracy. Mach Learn 51(2):181–207
Li G, Boukir S (2013) Margin-based ordered aggregation for ensemble pruning. Pattern Recogn Lett 34(6):603–609
Li N, Yu Y, Zhou Z-H (2012) Diversity regularized ensemble pruning. In: Machine learning and knowledge discovery in databases, pp 330–345
Lina C, Chena W, Qiua C, Wua Y, Krishnanc S, Zoua Q (2014) Libd3c: ensemble classifiers with a clustering and dynamic selection strategy. Neurocomputing 123:424–435
Liu P, Han S, Meng Z, Tong Y (2014) Facial expression recognition via a boosted deep belief network. In: 2014 IEEE conference on computer vision and pattern recognition, pp 1805–1812
Lopes AT, de Aguiar E, De Souza AF, Oliveira-Santos T (2017) Facial expression recognition with convolutional neural networks: coping with few data and the training sample order. Pattern Recogn 61:610–628
Lu Z, Wu X, Xingquan Z, Josh B (2010) Ensemble pruning via individual contribution ordering. In: KDD ’10 Proceedings of the 16th ACM SIGKDD international conference on knowledge discovery and data mining, pp 871–880
Lu ZY, Wu XD et al (2010) Ensemble pruning via individual contribution ordering. In: Presented at 16th ACM SIGKDD international conference on knowledge discovery and data mining, ACM, p 871–880
Lucey P et al (2010) The extended Cohn-Kanade dataset (CK + ): a complete expression dataset for action unit and emotion-specified expression. In: Proceedings third international workshop on CVPR for human communicative behavior analysis (CVPR4HB 2010), pp 94–101
Lyons M, Akamatsu S, Kamachi M, Gyoba J (1998) Coding facial expressions with Gabor wavelets. In: Third IEEE international conference on automatic face and gesture recognition, 1998. Proceedings, pp 200–205
Markatopouloub F, Tsoumakasa G, Vlahavasa I (2015) Dynamic ensemble pruning based on multi-label classification. Neurocomputing, Part B 150:501–512
Martinez-Munoz G, Suarez A (2004) Aggregation ordering in bagging. In: International conference on artificial intelligence and applications (IASTED), pp 258–263
Martinez-Munoz G, Suarez A (2006) Pruning in ordered bagging ensembles. In: 23rd international conference in machine learning (ICML 2006). ACM Press, pp 609–616
Martinez-Munoz G, Hernandez-Lobato D, Suarez A (2009) An analysis of ensemble pruning techniques based on ordered aggregation. IEEE Trans Pattern Anal Mach Intell 31(2):245–259
Mi A, Hai L (2010) A clustering-based classifier selection method for network intrusion detection. In: 2010 5th international conference on computer science and education (ICCSE) , pp 1001–1004
Michelle K, Wolfgang S, Jesse H, Dana K (2013) Human movement analysis: extension of the F-statistic to time series using HMM. In: 2013 IEEE international conference on systems, man, and cybernetics (SMC)
Moore S, Bowden R (2007) Automatic facial expression recognition using boosted discriminatory classifiers. In: AMFG’07 Proceedings of the 3rd international conference on analysis and modeling of faces and gestures, pp 71–83
Partalas I, Tsoumakas G, Vlahavas I (2009) Pruning an ensemble of classifiers via reinforcement learning. Neurocomputing 72(7–9):1900–1909
Peng H, Long F, Ding C (2005) Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy. IEEE Trans Pattern Anal Mach Intell 27(8):1226–1238
Tang Y (2013) Deep learning using linear support vector machines. In: Workshop on challenges in representation learning in ICML
Tamon C, Xiang J (2000) On the boosting pruning problem. In: Presented at 11th European conference on machine learning. Springer, Berlin, pp 404–412
Tsoumakas G, Partalas I, Vlahavas I (2009) Applications of supervised and unsupervised ensemble methods, vol 245. Springer, Berlin
Wang S, Chen K (2007) Ensemble learning with active data selection for semi-supervised pattern classification. In: 2007 international joint conference on neural networks, pp 355–360
Xiong L, Mao S, Jiao L (2015) Selective ensemble based on transformation of classifiers used SPCA. Int J Pattern Recognit Artif Intell 29(1). https://doi.org/10.1142/S0218001415500056
Zavaschi THH, Koerich AL, Oliveira LES (2011) Facial expression recognition using ensemble of classifiers. In: 2011 ieee international conference on acoustics, speech and signal processing (ICASSP)
Zhou ZH, Wu J, Tang W (2002) Ensembling neural networks: many could be better than all. Artif Intell 137(1-2):239–263
Zhuan L, Qun D, Ningzhong L (2014) Ensemble selection by GRASP. Appl Intell 41(1):128–144
Acknowledgments
This study was supported by a China National Science Foundation under Grants (60973083, 61273363), Science and Technology Planning Project of Guangdong Province (2014A010103009, 2015A020217002), and Guangzhou Science and Technology Planning Project (201504291154480).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Li, D., Wen, G. MRMR-based ensemble pruning for facial expression recognition. Multimed Tools Appl 77, 15251–15272 (2018). https://doi.org/10.1007/s11042-017-5105-z
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-017-5105-z